Skip to content
WorkshopOnline

Zooming in on a Forecast

Event artwork for Zooming in on a Forecast

Supporting this event

Sponsors

dunnhumby
About this event

What to expect

Learn how to add granular detail to large-scale retail forecasts without rebuilding existing forecasting systems in this free online DSF Sandbox Session.

Thursday 26 November 2026, 1:00–2:00 pm GMT · Online webinar

Marco Brambilla, Research Data Science Specialist at dunnhumby, will examine a recurring retail analytics challenge: stakeholders want forecasts segmented by shopper type, demographics or other business dimensions, but producing more granular models can be costly and complex.

Using photography as an analogy, Marco will compare three ways to “zoom in” on a forecast:

  • Optical zoom: building models that capture genuine additional detail
  • Digital zoom: using simpler methods such as historical averages
  • AI-enhanced zoom: creating more detailed forecasts without replacing the underlying system

Attendees will gain a practical framework for deciding when simple techniques are sufficient, when further modelling is justified and how predictive detail can be added at scale.

Marco brings 18 years of experience at dunnhumby, where he has designed automated demand-forecasting solutions for major grocery retailers and led work spanning pricing, forecasting and intervention measurement. His interests include mathematical modelling, statistical inference and causal reasoning.

This session is part of Data Science Festival Bristol’s free monthly Sandbox Sessions. Meetup attendance alone does not register you. Complete the official registration form to receive your unique webinar link and email reminders.

View the event listing and registration instructions

Running an AI event?

List it here, free.

Meetups, workshops and conferences go in front of thousands of builders after a quick review.

List your event